CN112527821A - Student bloom mastery degree evaluation method, system and storage medium - Google Patents
Student bloom mastery degree evaluation method, system and storage medium Download PDFInfo
- Publication number
- CN112527821A CN112527821A CN202011449802.2A CN202011449802A CN112527821A CN 112527821 A CN112527821 A CN 112527821A CN 202011449802 A CN202011449802 A CN 202011449802A CN 112527821 A CN112527821 A CN 112527821A
- Authority
- CN
- China
- Prior art keywords
- student
- bloom
- state
- unit
- lstm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 40
- 230000006399 behavior Effects 0.000 claims abstract description 104
- 238000000034 method Methods 0.000 claims abstract description 43
- 238000012549 training Methods 0.000 claims abstract description 39
- 238000012360 testing method Methods 0.000 claims description 28
- 238000011835 investigation Methods 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 18
- 238000010606 normalization Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000011478 gradient descent method Methods 0.000 claims description 6
- 230000003542 behavioural effect Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 abstract description 6
- 230000006998 cognitive state Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 15
- 238000010801 machine learning Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000019771 cognition Effects 0.000 description 3
- 230000001149 cognitive effect Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B25/00—Models for purposes not provided for in G09B23/00, e.g. full-sized devices for demonstration purposes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a student brucm mastery degree evaluation method, a system and a storage medium, wherein the method comprises the following steps: constructing and training a depth knowledge tracking model; when a new learning behavior is generated by learning, predicting the correct rate of doing exercises of students on each exercise through the trained deep knowledge tracking model; and evaluating the knowledge point Broumu mastery degree of the student at the current moment based on the prediction result of the question making accuracy of the student on each question. The method can model the brucm mastery degree of the knowledge points through the learning behavior sequence and the deep knowledge tracking method of the students, visually represent the ability information and the mastery degree of the students on the granularity of the knowledge points, can be deployed in an online education and mixed education platform, effectively represent the learning ability change and the cognitive state change of the students, and achieves the purpose of brucm operation.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a student bloom mastery degree evaluation method, a student bloom mastery degree evaluation system and a storage medium.
Background
The bruhm cognitive model is a classification method proposed by the american psychologist of education, benjiming bruhm, which classifies the educational objectives of educators so as to effectively achieve each objective. The bruhm cognitive model comprises six grades, which are respectively: know, understand, apply, analyze, synthesize, evaluate. The bloom theory can effectively guide the teaching process, and reasonably describe the cognitive process of students, but a large amount of expert knowledge is needed. Because a large amount of structured and unstructured learning behavior data can be generated in the fields of online education and mixed improvement education, how to intelligently model the knowledge point Broumu mastery of students faces a dilemma, and the Broumu model theory is difficult to fall on the ground in an application scene.
Disclosure of Invention
The invention provides a student bloom mastery degree evaluation method, a student bloom mastery degree evaluation system and a storage medium. The bloom operation is a conversion calculation from the learning behavior of a student to the bloom cognition level reached by the student, the input of the bloom operation is a learning behavior sequence of the student, and the output of the bloom operation is the bloom cognition level corresponding to the learning effect.
In order to achieve the purpose of the invention, the invention provides the following technical scheme:
the invention provides a student brucm mastery degree evaluation method, which comprises the following steps:
constructing and training a depth knowledge tracking model; the deep knowledge tracking model is obtained by modeling potential relations among student abilities, test question difficulty and question making accuracy, and is trained through a learning behavior sequence of students;
when new student behaviors are produced by learning, the correct rate of doing exercises on each exercise of the student is predicted through the trained deep knowledge tracking model;
determining student identity marks and knowledge points to be evaluated;
inquiring all teaching resources from the relational database, and extracting all exercises for inspecting the knowledge points;
grouping the test questions according to the bloom investigation level of the test questions;
normalizing the difficulty of investigation of all the problems in each group;
matching the extracted exercises with the depth knowledge tracking model to the prediction probability of each exercise answered by the student, and weighting according to the normalized investigation difficulty; taking the weighting result as the bloom score of the knowledge point under the corresponding bloom grade;
and setting a threshold value for the bloom mastery degree under each bloom grade, and evaluating the bloom grade reached by the student at the knowledge point according to the bloom score.
Further, constructing and training a deep knowledge tracking model, comprising:
constructing an LSTM network, and initializing the structural parameters of the network;
reading learning behavior information of students, including exercise information and learning results, and coding the learning behaviors of the students by using a one-bit effective coding method to form a learning behavior embedded layer; the length of the learning behavior embedding layer is 2N, and N is the total number of the questions in the question bank;
reading the state of the LSTM hidden layer at the last moment; if the device is in a cold start stage, initializing the state of the hidden layer; the LSTM hidden layer states include: LSTM cell state values and LSTM hidden state values;
inputting the learning behavior embedding layer and the previous time hidden layer state into an LSTM network, and calculating and storing the current time LSTM hidden layer state;
calculating a loss function, and updating LSTM network parameters by using a gradient descent method; the loss function is:
wherein r istIs the real learning behavior result of the student at the time t, ptIs the probability of the model predicting the behavioral outcome;
and repeating the steps until the training of all the learning behavior data is completed.
Further, after completing the training of all the learning behavior data, the method further includes: and storing the trained deep knowledge tracking model into a relational database.
Further, when a student learns to generate new student behaviors, the correct rate of the student doing exercises on each exercise is predicted through the trained deep knowledge tracking model, and the method comprises the following steps:
loading a trained depth knowledge tracking model;
obtaining the LSTM hidden layer state at the last moment from a relational database according to the identity of the student;
carrying out one-bit effective coding on the new learning behavior information generated by the student to form a learning behavior embedding layer at the current moment;
inputting the LSTM hidden layer state at the last moment and the learning behavior embedding layer at the current moment into a loaded deep knowledge tracking model, calculating the LSTM hidden layer state at the current moment according to the trained network structure parameters, and storing the LSTM hidden layer state at the current moment into a relational database;
according to pt＝σ(W[ht-1,xt]+ b) calculating the prediction result of the student question making accuracy at the current moment, wherein ht-1Hidden layer states at time t-1, xtFor the input layer state at time t, W and b are the weight coefficient and bias coefficient, σ (·) of the training network, respectively) Is an activation function.
The invention also provides a student brucm mastery degree evaluation system, which comprises the following steps:
the depth knowledge tracking model training module is used for constructing and training a depth knowledge tracking model; the deep knowledge tracking model is obtained by modeling potential relations among student abilities, test question difficulty and question making accuracy, and is trained through a learning behavior sequence of students;
the student exercise correct rate prediction module is used for predicting the exercise correct rate of students on each exercise through the deep knowledge tracking model trained by the deep knowledge tracking model training module when new student behaviors are generated;
the knowledge point bloom mastery degree evaluation module is used for evaluating the knowledge point bloom mastery degree of the student at the current moment based on the prediction result of the student question making correctness of the student on each question obtained by the student question making correctness prediction module;
wherein, knowledge point brucm mastery degree evaluation module includes:
the determination unit is used for determining the identity of the student and the knowledge point to be evaluated;
the extraction unit is used for inquiring all teaching resources from the relational database and extracting all exercises for inspecting the knowledge points;
the grouping unit is used for grouping the test questions extracted by the extraction unit according to the Blum investigation level of the exercise questions;
the normalization unit is used for normalizing the investigation difficulty of all the exercises in each group obtained by the grouping unit;
the evaluation unit is used for matching the extracted exercises with the prediction probability of each exercise given by the student and normalized by the normalization unit according to the depth knowledge tracking model, and weighting the evaluation according to the normalized investigation difficulty of the normalization unit; taking the weighting result as the bloom score of the knowledge point under the corresponding bloom grade;
and the evaluation unit is used for setting a threshold value for the bloom mastery degree under each bloom grade, and evaluating the bloom grade reached by the student on the knowledge point according to the bloom grade obtained by the grading unit.
Further, the deep knowledge tracking model training module comprises:
the construction unit is used for constructing the LSTM network and initializing the structural parameters of the network;
the first coding unit is used for reading the learning behavior information of students, including exercise information and learning results, and coding the learning behaviors of the students by using a one-bit effective coding method to form a learning behavior embedded layer; the length of the learning behavior embedding layer is 2N, and N is the total number of the questions in the question bank;
a first state reading unit for reading the state of the LSTM hidden layer at the last moment; if the device is in a cold start stage, initializing the state of the hidden layer; the LSTM hidden layer states include: LSTM cell state values and LSTM hidden state values;
the first state calculating unit is used for inputting the learning behavior embedded layer obtained by the first coding unit and the hidden layer state of the last moment obtained by the first state reading unit into the LSTM network, and calculating and storing the LSTM hidden layer state of the current moment;
the first updating network parameter unit is used for calculating a loss function and updating the LSTM network parameter by using a gradient descent method; the loss function is:
wherein r istIs the real learning behavior result of the student at the time t, ptIs the probability of the model predicting the behavioral outcome;
and the storage unit is used for storing the trained deep knowledge tracking model into the relational database after finishing the training of all the learning behavior data.
Further, the student question making accuracy prediction module comprises:
the loading unit is used for loading the trained depth knowledge tracking model;
the second state reading unit is used for acquiring the LSTM hidden layer state at the last moment from the relational database according to the student identity;
the second coding unit is used for carrying out one-bit effective coding on the new learning behavior information generated by the student to form a learning behavior embedding layer at the current moment;
the second state calculating unit is used for inputting the LSTM hidden layer state at the last moment obtained by the second state reading unit and the learning behavior embedding layer obtained by the second coding unit into the loaded deep knowledge tracking model, calculating the LSTM hidden layer state at the current moment according to the trained network structure parameters, and storing the LSTM hidden layer state into the relational database;
a prediction unit for predicting according to pt＝σ(W[ht-1,xt]+ b) calculating the prediction result of the student question making accuracy at the current moment, wherein ht-1Hidden layer states at time t-1, xtFor the input layer state at time t, W and b are the weight coefficient and bias coefficient of the training network, respectively, and σ (-) is the activation function.
The invention also provides a computer readable storage medium, wherein a computer instruction set is stored in the computer readable storage medium, and when being executed by a processor, the computer instruction set realizes the student bloom mastery degree evaluation method provided by the invention.
The invention has the advantages and positive effects that:
the method can model the brucm mastery degree of the knowledge points through the learning behavior sequence and the deep knowledge tracking method of the students, visually represent the ability information and the mastery degree of the students on the granularity of the knowledge points, can be deployed in an online education and mixed education platform, effectively represent the learning ability change and the cognitive state change of the students, and achieves the purpose of brucm operation.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flowchart of a method for evaluating the degree of mastery of a student Broomu according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating the process of building and training a depth knowledge tracking model according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating how to predict the correct rate of student questions made in the embodiment of the present invention;
FIG. 4 is a process of evaluating the degree of mastery of the student knowledge point Broumu in the embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Students generate a large amount of learning data in the teaching platform, wherein the learning data comprises information such as learning resources and learning results, and the data reflects the potential mastery degree of the students on each knowledge point. Therefore, the learning behavior data can be encoded and input into an LSTM (Long Short Term Memory networks) model, and the accuracy of student question making is predicted by using a deep knowledge tracking method. Furthermore, as the teaching resources comprise the inspected knowledge point information and the bloom inspection grade, the accuracy of student response reflects bloom mastery degree information of the corresponding knowledge point, so that the bloom mastery degree of the knowledge point can be modeled.
Referring to fig. 1, a flowchart of a method for evaluating bloom mastery of a student in the embodiment of the present invention is shown, where the method includes deep knowledge tracking model construction and training, student question making accuracy prediction, and knowledge point bloom mastery evaluation.
The deep knowledge tracking model is trained by modeling potential relations among student abilities, test question difficulty and question making accuracy through a large number of student learning behavior sequences, when a new learning behavior is generated, the question making accuracy of students on all questions can be predicted through the trained model, and the knowledge point Brohum mastery of the students at the current moment can be further evaluated by utilizing a Brohum operation matrix representing question-knowledge point association, so that the purpose of Brohum operation is achieved.
1. Depth knowledge tracking model construction and training
The deep knowledge tracking method inputs a student learning behavior sequence into a cyclic neural network, represents a potential ability model of a student through a hidden layer state, and simultaneously outputs a prediction result of the correct rate of student question making. And (4) training the deep learning model by taking the real question making result as a label and the cross entropy as a loss function. The specific training process is shown in fig. 2, and comprises the following steps:
s101, constructing an LSTM network, and initializing the structural parameters of the network.
S102, reading learning behavior information of the students, including exercise information and learning results, and coding the learning behaviors of the students by using a one-bit effective coding (one-hot coding) method to obtain a learning behavior Embedding layer (Embedding); wherein, the length of the learning behavior Embedding is 2N, and N is the total number of the problem bank problems.
S103, reading the LSTM hidden layer state at the last time (t-1 time), including: last time LSTM cell state value (LSTM-C)t-1) And last time LSTM hidden state value (LSTM-H)t-1). If the device is in a cold start stage, the unit state and the hidden state need to be initialized.
In the Tensorflow deep learning framework, the hidden layer states can be read and saved by setting them to Variable variables. Other frames such as Pythrch, etc.
S104, learning behaviors Embedding and LSTM-Ct-1And LSTM-Ht-1Inputting the LSTM into an LSTM network, calculating and storing the LSTM hidden layer state at the current moment, wherein the method comprises the following steps: the LSTM unit state value (LSTM-Ct) at the current moment and the LSTM hidden state value (LSTM-Ht) at the current moment.
S105, calculating a loss function shown in the formula 1, and updating LSTM network parameters by using a gradient descent method:
wherein r istIs the real learning behavior result of the student at the time t, ptIs the probability that the model predicts the outcome of the behavior.
And S106, repeating S102-S105 until the training of all the behavior data is completed.
Then also comprises the following steps: and storing the trained deep knowledge tracking model in a relational database.
The process of constructing and training the depth knowledge tracking model can fully utilize the learning behavior sequence information generated by students in the learning process, and the capability characteristics of the students are mined from the student behavior information through the depth knowledge tracking model.
2. Student question making accuracy prediction
The online education platform and the mixed education platform have the requirement of real-time updating on the Brumm level of the student knowledge point, so that the purpose of synchronization of mastery evaluation and student learning behaviors is achieved. Therefore, the trained deep knowledge tracking model needs to be deployed on line, and the capability model, namely the question correctness prediction, needs to be updated in real time according to the learning behavior of the student on the learning platform. The specific prediction process is shown in fig. 3, and includes the following steps:
s201, loading a depth knowledge tracking model.
S202, obtaining the LSTM hidden layer state at the last moment from the relational database according to the student identity (id).
And S203, performing one-hot coding on the learning behavior data of the student at the current moment to form learning behavior Embedding.
S204, inputting the LSTM hidden layer state at the last moment and the learning behavior Embedding at the current moment into the loaded LSTM network, calculating the hidden layer state at the current moment according to the trained network structure parameters, and storing the hidden layer state at the current moment into a relational database.
S205, calculating a prediction result of the student question making accuracy at the current moment according to the formula 2;
pt＝σ(W[ht-1,xt]+b) (2)
wherein h ist-1Hidden layer states at time t-1, xtFor the input layer state at time t, W and b are the weight coefficient and bias coefficient of the training network, respectively, and σ (-) is the activation function.
According to the student question making accuracy prediction process, the off-line training deep knowledge tracking model is deployed in the on-line learning platform, the ability characteristics of the student can be dynamically updated according to the new learning behavior of the student in the learning platform in real time, and the student question making accuracy is predicted in real time.
3. Knowledge point brucm mastery evaluation
The student correct rate of doing exercises can characterize the comprehensive ability of the student from the angle of exercise test, but the degree of mastering of the student on each knowledge point cannot be expressed visually, and especially when a plurality of knowledge points are investigated simultaneously and the investigation difficulty is different, the degree of mastering of the student on the knowledge points is difficult to embody. Therefore, the student question making accuracy information needs to be further converted into the knowledge point broomm mastery degree, and a specific evaluation method is shown in fig. 4 and comprises the following steps:
s301, determining student id and knowledge points to be evaluated.
S302, inquiring all teaching resources from a relational database, and extracting all exercises for inspecting the knowledge points; on the basis, the test questions are grouped according to the Blum investigation level of the test questions.
And S303, normalizing the investigation difficulty of all the exercises in each group.
S304, reading the prediction probability of the deep knowledge tracking model for each exercise answered by the student from the relational database, matching the exercise extracted in S302 with the corresponding prediction probability, and weighting according to the normalized investigation difficulty in S303; and taking the weighted result as the bloom score of the knowledge point under the corresponding bloom grade.
S305, setting a threshold value for the bloom mastery degree under each bloom grade, and further evaluating the bloom grade reached by the student at the knowledge point according to the bloom score.
According to the knowledge point bloom mastery degree evaluation method, relatively abstract student question making result prediction information is converted into knowledge point bloom grade information, the mastery degree of each knowledge point by students is mined and evaluated from the prediction result output by the model, and bloom operation is performed on the students from the aspects of capability characteristics, teaching resources and behavior results.
The embodiment of the invention can model the bloom mastery degree of the knowledge point through the learning behavior sequence and deep knowledge tracking method of the student, visually represent the capability information and the mastery degree of the student on the granularity of the knowledge point, can be deployed in an online education and mixed education platform, effectively represent the learning capability change and the cognitive state change of the student and achieve the goal of bloom operation.
In order to facilitate understanding of the overall process of bloom operation in the present invention, a real application process of machine learning in an online teaching platform is described as an example.
The specific application process comprises the following three links:
first, depth knowledge tracking model training
A large number of learning behavior records of students in the past period are stored in the online teaching platform, the learning behavior data in the machine learning course is preliminarily cleaned, and the learning behavior data is supplemented by the simulation behavior data. The final application contained approximately 1,800,000 learning behavior records generated by 5000 students in the data set trained by the deep knowledge tracking model. The student learning behavior data structure is shown in table 1, for example.
TABLE 1
| Record id | Student id | Test question id | Behavioral results | Total number of actions |
| 6137 | 89 | 244 | 0 | 1 |
| 6138 | 89 | 523 | 1 | 2 |
| 6139 | 89 | 29 | 0 | 3 |
| 6140 | 89 | 229 | 0 | 4 |
| 6141 | 89 | 329 | 1 | 5 |
| 6142 | 89 | 329 | 0 | 6 |
| 6143 | 89 | 245 | 1 | 7 |
| 6144 | 89 | 520 | 1 | 8 |
| 6145 | 89 | 420 | 0 | 9 |
| 6146 | 89 | 320 | 0 | 10 |
| 6147 | 89 | 256 | 0 | 11 |
| 6148 | 89 | 223 | 1 | 12 |
| 6149 | 89 | 409 | 1 | 13 |
The data set is divided according to the proportion of 7:2:1, and a deep knowledge tracking model is trained, wherein the accuracy of the trained model on the verification set is 72.1%, and the AUC is 77.2%. (Note: this link model is the most basic structure in view of the on-line deployment and real-time feedback requirements of the model.)
Second, student question-making accuracy assessment
Taking student A who is participating in learning activities in a teaching platform as an example, past learning behavior records are stored in the platform, and new behavior data can be continuously generated in the learning process.
The teaching platform periodically calls the bloom operation service, and the module reads newly generated learning behavior data of the student A from the database and processes the learning behavior data into serialized data consisting of learning behavior Embedding. And calling a deep knowledge tracking model to evaluate the correctness of the questions in the current state on the basis of reading the state of the current hidden layer. The question bank question data structure is shown in table 2.
TABLE 2
Since the problem library of the "machine learning" course contains 1573 test questions, the final evaluation result is a vector with a length of 1573, and the vector contains the probability of the student a for each test question in the current state.
P＝[0.87,0.24,0.57,......0.42,0.23,0.92,0.85,......0.25,0.85,0.78,0.91]1×1573
Third, evaluation of Broumu mastery
The bloom mastery degree takes the knowledge point as the minimum granularity, so that a bloom operation matrix is required to convert the test question accuracy rate into the bloom mastery degree. The course of machine learning has 83 knowledge points, and all 1573 test questions in the question bank are marked with investigation knowledge points, investigation difficulty and bloom investigation level of the test questions through an algorithm or manually. Table 3 shows the examination relation between the test question and the knowledge point
TABLE 3
| Test question id | Investigating knowledge points | Difficulty of investigation | Blum investigation class |
| 51 | Convolutional neural network | 0.333 | 2 |
| 53 | Convolutional neural network | 0.667 | 2 |
| 54 | Convolutional neural network | 1.000 | 2 |
For each knowledge point, firstly matching all test questions for inspecting the knowledge point in a question bank, and classifying according to the Blume inspection grade. In each bloom classification, the investigation difficulty of all test questions in the class is normalized, the normalized investigation difficulty is multiplied by the probability of the test questions, and the accumulated result is the evaluation score of the knowledge point in the bloom class.
Taking the bloom secondary evaluation of the convolutional neural network as an example, extracting a test question list from the question bank as [51, 53, 54], normalizing the difficulty coefficient to [0.167, 0.333, 0.500], matching the current question making accuracy rate of student a as [0.42, 0.92, 0.85], and then evaluating the result as:
Score＝0.42*0.167+0.92*0.333+0.85*0.500＝0.802；
and evaluating the bloom cognitive models of all knowledge points in the course to obtain a bloom mastery evaluation result of the student A in the current state for the machine learning course with the knowledge points as granularity.
The invention also provides a student bloom mastery degree evaluation system, which corresponds to the student bloom mastery degree evaluation method and comprises the following steps:
the depth knowledge tracking model training module is used for constructing and training a depth knowledge tracking model; the deep knowledge tracking model is obtained by modeling potential relations among student abilities, test question difficulty and question making accuracy, and is trained through a learning behavior sequence of students;
the student exercise correct rate prediction module is used for predicting the exercise correct rate of students on each exercise through the deep knowledge tracking model trained by the deep knowledge tracking model training module when new student behaviors are generated;
the knowledge point bloom mastery degree evaluation module is used for evaluating the knowledge point bloom mastery degree of the student at the current moment based on the prediction result of the student question making correctness of the student on each question obtained by the student question making correctness prediction module;
wherein, knowledge point brucm mastery degree evaluation module includes:
the determination unit is used for determining the identity of the student and the knowledge point to be evaluated;
the extraction unit is used for inquiring all teaching resources from the relational database and extracting all exercises for inspecting the knowledge points;
the grouping unit is used for grouping the test questions extracted by the extraction unit according to the Blum investigation level of the exercise questions;
the normalization unit is used for normalizing the investigation difficulty of all the exercises in each group obtained by the grouping unit;
the evaluation unit is used for matching the extracted exercises with the prediction probability of each exercise given by the student and normalized by the normalization unit according to the depth knowledge tracking model, and weighting the evaluation according to the normalized investigation difficulty of the normalization unit; taking the weighting result as the bloom score of the knowledge point under the corresponding bloom grade;
and the evaluation unit is used for setting a threshold value for the bloom mastery degree under each bloom grade, and evaluating the bloom grade reached by the student on the knowledge point according to the bloom grade obtained by the grading unit.
Further, the deep knowledge tracking model training module comprises:
the construction unit is used for constructing the LSTM network and initializing the structural parameters of the network;
the first coding unit is used for reading the learning behavior information of students, including exercise information and learning results, and coding the learning behaviors of the students by using a one-bit effective coding method to form a learning behavior embedded layer; the length of the learning behavior embedding layer is 2N, and N is the total number of the questions in the question bank;
a first state reading unit for reading the state of the LSTM hidden layer at the last moment; if the device is in a cold start stage, initializing the state of the hidden layer; the LSTM hidden layer states include: LSTM cell state values and LSTM hidden state values;
the first state calculating unit is used for inputting the learning behavior embedded layer obtained by the first coding unit and the hidden layer state of the last moment obtained by the first state reading unit into the LSTM network, and calculating and storing the LSTM hidden layer state of the current moment;
the first updating network parameter unit is used for calculating a loss function and updating the LSTM network parameter by using a gradient descent method; the loss function is:
wherein r istIs the real learning behavior result of the student at the time t, ptIs the probability that the model predicts the outcome of the behavior.
And the storage unit is used for storing the trained deep knowledge tracking model into the relational database after finishing the training of all the learning behavior data.
Further, the student question making accuracy prediction module comprises:
the loading unit is used for loading the trained depth knowledge tracking model;
the second state reading unit is used for acquiring the LSTM hidden layer state at the last moment from the relational database according to the student identity;
the second coding unit is used for carrying out one-bit effective coding on the new learning behavior information generated by the student to form a learning behavior embedding layer at the current moment;
the second state calculating unit is used for inputting the LSTM hidden layer state at the last moment obtained by the second state reading unit and the learning behavior embedding layer obtained by the second coding unit into the loaded deep knowledge tracking model, calculating the LSTM hidden layer state at the current moment according to the trained network structure parameters, and storing the LSTM hidden layer state into the relational database;
a prediction unit for predicting according to pt＝σ(W[ht-1,xt]+ b) calculating the prediction result of the student question making accuracy at the current moment, wherein ht-1Hidden layer states at time t-1, xtFor the input layer state at time t, W and b are the weight coefficient and bias coefficient of the training network, respectively, and σ (-) is the activation function.
For the embodiments of the present invention, the description is simple because it corresponds to the above embodiments, and for the related similarities, please refer to the description in the above embodiments, and the detailed description is omitted here.
The embodiment of the invention also discloses a computer-readable storage medium, wherein a computer instruction set is stored in the computer-readable storage medium, and when being executed by a processor, the computer instruction set realizes the student bloom mastery degree evaluation method provided by any one of the above embodiments.
In the embodiments provided in the present invention, it should be understood that the disclosed technical contents can be implemented in other manners. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A student bloom mastery degree evaluation method is characterized by comprising the following steps:
constructing and training a depth knowledge tracking model; the deep knowledge tracking model is obtained by modeling potential relations among student abilities, test question difficulty and question making accuracy, and is trained through a learning behavior sequence of students;
when new student behaviors are produced by learning, the correct rate of doing exercises on each exercise of the student is predicted through the trained deep knowledge tracking model;
determining student identity marks and knowledge points to be evaluated;
inquiring all teaching resources from the relational database, and extracting all exercises for inspecting the knowledge points;
grouping the test questions according to the bloom investigation level of the test questions;
normalizing the difficulty of investigation of all the problems in each group;
matching the extracted exercises with the depth knowledge tracking model to the prediction probability of each exercise answered by the student, and weighting according to the normalized investigation difficulty; taking the weighting result as the bloom score of the knowledge point under the corresponding bloom grade;
and setting a threshold value for the bloom mastery degree under each bloom grade, and evaluating the bloom grade reached by the student at the knowledge point according to the bloom score.
2. The method of claim 1, wherein constructing and training a deep knowledge tracking model comprises:
constructing an LSTM network, and initializing the structural parameters of the network;
reading learning behavior information of students, including exercise information and learning results, and coding the learning behaviors of the students by using a one-bit effective coding method to form a learning behavior embedded layer; the length of the learning behavior embedding layer is 2N, and N is the total number of the questions in the question bank;
reading the state of the LSTM hidden layer at the last moment; if the device is in a cold start stage, initializing the state of the hidden layer; the LSTM hidden layer states include: LSTM cell state values and LSTM hidden state values;
inputting the learning behavior embedding layer and the previous time hidden layer state into an LSTM network, and calculating and storing the current time LSTM hidden layer state;
calculating a loss function, and updating LSTM network parameters by using a gradient descent method; the loss function is:
wherein r istIs the real learning behavior result of the student at the time t, ptIs the probability of the model predicting the behavioral outcome;
and repeating the steps until the training of all the learning behavior data is completed.
3. The method of claim 2, wherein after completing training of all learning behavior data, further comprising: and storing the trained deep knowledge tracking model into a relational database.
4. The method of claim 3, wherein when learning to generate new student behaviors, predicting the correct rate of student doing exercises on each exercise through the trained deep knowledge tracking model comprises:
loading a trained depth knowledge tracking model;
obtaining the LSTM hidden layer state at the last moment from a relational database according to the identity of the student;
carrying out one-bit effective coding on the new learning behavior information generated by the student to form a learning behavior embedding layer at the current moment;
inputting the LSTM hidden layer state at the last moment and the learning behavior embedding layer at the current moment into a loaded deep knowledge tracking model, calculating the LSTM hidden layer state at the current moment according to the trained network structure parameters, and storing the LSTM hidden layer state at the current moment into a relational database;
according to pt＝σ(W[ht-1,xt]+ b) calculating the prediction result of the student question making accuracy at the current moment, wherein ht-1Hidden layer states at time t-1, xtFor the input layer state at time t, W and b are the weight coefficient and bias coefficient of the training network, respectively, and σ (-) is the activation function.
5. A student bloom mastery degree evaluation system is characterized by comprising:
the depth knowledge tracking model training module is used for constructing and training a depth knowledge tracking model; the deep knowledge tracking model is obtained by modeling potential relations among student abilities, test question difficulty and question making accuracy, and is trained through a learning behavior sequence of students;
the student exercise correct rate prediction module is used for predicting the exercise correct rate of students on each exercise through the deep knowledge tracking model trained by the deep knowledge tracking model training module when new learning behaviors are generated;
the knowledge point bloom mastery degree evaluation module is used for evaluating the knowledge point bloom mastery degree of the student at the current moment based on the prediction result of the student question making correctness of the student on each question obtained by the student question making correctness prediction module;
wherein, knowledge point brucm mastery degree evaluation module includes:
the determination unit is used for determining the identity of the student and the knowledge point to be evaluated;
the extraction unit is used for inquiring all teaching resources from the relational database and extracting all exercises for inspecting the knowledge points;
the grouping unit is used for grouping the test questions extracted by the extraction unit according to the Blum investigation level of the exercise questions;
the normalization unit is used for normalizing the investigation difficulty of all the exercises in each group obtained by the grouping unit;
the evaluation unit is used for matching the extracted exercises with the prediction probability of each exercise given by the student and normalized by the normalization unit according to the depth knowledge tracking model, and weighting the evaluation according to the normalized investigation difficulty of the normalization unit; taking the weighting result as the bloom score of the knowledge point under the corresponding bloom grade;
and the evaluation unit is used for setting a threshold value for the bloom mastery degree under each bloom grade, and evaluating the bloom grade reached by the student on the knowledge point according to the bloom grade obtained by the grading unit.
6. The system of claim 5, wherein the deep knowledge tracking model training module comprises:
the construction unit is used for constructing the LSTM network and initializing the structural parameters of the network;
the first coding unit is used for reading the learning behavior information of students, including exercise information and learning results, and coding the learning behaviors of the students by using a one-bit effective coding method to form a learning behavior embedded layer; the length of the learning behavior embedding layer is 2N, and N is the total number of the questions in the question bank;
a first state reading unit for reading the state of the LSTM hidden layer at the last moment; if the device is in a cold start stage, initializing the state of the hidden layer; the LSTM hidden layer states include: LSTM cell state values and LSTM hidden state values;
the first state calculating unit is used for inputting the learning behavior embedded layer obtained by the first coding unit and the hidden layer state of the last moment obtained by the first state reading unit into the LSTM network, and calculating and storing the LSTM hidden layer state of the current moment;
the first updating network parameter unit is used for calculating a loss function and updating the LSTM network parameter by using a gradient descent method; the loss function is:
wherein r istIs the real learning behavior result of the student at the time t, ptIs the probability of the model predicting the behavioral outcome;
and the storage unit is used for storing the trained deep knowledge tracking model into the relational database after finishing the training of all the learning behavior data.
7. The system of claim 6, wherein the student question making accuracy prediction module comprises:
the loading unit is used for loading the trained depth knowledge tracking model;
the second state reading unit is used for acquiring the LSTM hidden layer state at the last moment from the relational database according to the student identity;
the second coding unit is used for carrying out one-bit effective coding on the new learning behavior information generated by the student to form a learning behavior embedding layer at the current moment;
the second state calculating unit is used for inputting the LSTM hidden layer state at the last moment obtained by the second state reading unit and the learning behavior embedding layer obtained by the second coding unit into the loaded deep knowledge tracking model, calculating the LSTM hidden layer state at the current moment according to the trained network structure parameters, and storing the LSTM hidden layer state into the relational database;
a prediction unit for predicting according to pt＝σ(W[ht-1,xt]+ b) calculating the prediction result of the student question making accuracy at the current moment, wherein ht-1Hidden layer states at time t-1, xtFor the input layer state at time t, W and b are the weight coefficient and bias coefficient of the training network, respectively, and σ (-) is the activation function.
8. A computer-readable storage medium having stored thereon a set of computer instructions which, when executed by a processor, implement the student blume mastery assessment method as provided in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011449802.2A CN112527821A (en) | 2020-12-09 | 2020-12-09 | Student bloom mastery degree evaluation method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011449802.2A CN112527821A (en) | 2020-12-09 | 2020-12-09 | Student bloom mastery degree evaluation method, system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112527821A true CN112527821A (en) | 2021-03-19 |
Family
ID=74998752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011449802.2A Pending CN112527821A (en) | 2020-12-09 | 2020-12-09 | Student bloom mastery degree evaluation method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112527821A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113157372A (en) * | 2021-04-27 | 2021-07-23 | 万里安飞行服务(上海)有限公司 | Electronic book dynamic identification method based on knowledge mastery degree |
| CN113220847A (en) * | 2021-06-01 | 2021-08-06 | 平安科技（深圳）有限公司 | Knowledge mastering degree evaluation method and device based on neural network and related equipment |
| CN113743083A (en) * | 2021-09-06 | 2021-12-03 | 东北师范大学 | Test question difficulty prediction method and system based on deep semantic representation |
| CN114238613A (en) * | 2022-02-22 | 2022-03-25 | 北京一起航帆科技有限公司 | Method and device for determining mastery degree of knowledge points and electronic equipment |
| CN114529436A (en) * | 2022-02-10 | 2022-05-24 | 珠海读书郎软件科技有限公司 | Knowledge point mastery degree evaluation method, device and medium |
| CN116308935A (en) * | 2023-05-19 | 2023-06-23 | 湖南师范大学 | Learning early warning method and device based on multi-feature modeling and multi-level evaluation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032651A (en) * | 2019-04-18 | 2019-07-19 | 江苏师范大学 | A kind of constructing method of the individualized learning characteristic model of knowledge based map |
| CN110851723A (en) * | 2019-11-14 | 2020-02-28 | 上海钦文信息科技有限公司 | English exercise recommendation method based on large-scale knowledge point labeling result |
| CN110930274A (en) * | 2019-12-02 | 2020-03-27 | 中山大学 | Practice effect evaluation and learning path recommendation system and method based on cognitive diagnosis |
| CN110991645A (en) * | 2019-11-18 | 2020-04-10 | 广东宜学通教育科技有限公司 | Self-adaptive learning method, system and storage medium based on knowledge model |
-
2020
- 2020-12-09 CN CN202011449802.2A patent/CN112527821A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032651A (en) * | 2019-04-18 | 2019-07-19 | 江苏师范大学 | A kind of constructing method of the individualized learning characteristic model of knowledge based map |
| CN110851723A (en) * | 2019-11-14 | 2020-02-28 | 上海钦文信息科技有限公司 | English exercise recommendation method based on large-scale knowledge point labeling result |
| CN110991645A (en) * | 2019-11-18 | 2020-04-10 | 广东宜学通教育科技有限公司 | Self-adaptive learning method, system and storage medium based on knowledge model |
| CN110930274A (en) * | 2019-12-02 | 2020-03-27 | 中山大学 | Practice effect evaluation and learning path recommendation system and method based on cognitive diagnosis |
Non-Patent Citations (1)
| Title |
|---|
| 谭明新等: ""基于用户模型的生成性学习资源个性化推荐研究"", 《软件导刊》, vol. 16, no. 8, 30 August 2017 (2017-08-30), pages 123 - 127 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113157372A (en) * | 2021-04-27 | 2021-07-23 | 万里安飞行服务(上海)有限公司 | Electronic book dynamic identification method based on knowledge mastery degree |
| CN113157372B (en) * | 2021-04-27 | 2024-02-02 | 万里安飞行服务(上海)有限公司 | Electronic book dynamic identification method based on knowledge mastering degree |
| CN113220847A (en) * | 2021-06-01 | 2021-08-06 | 平安科技（深圳）有限公司 | Knowledge mastering degree evaluation method and device based on neural network and related equipment |
| CN113743083A (en) * | 2021-09-06 | 2021-12-03 | 东北师范大学 | Test question difficulty prediction method and system based on deep semantic representation |
| CN113743083B (en) * | 2021-09-06 | 2024-03-12 | 东北师范大学 | Test question difficulty prediction method and system based on deep semantic characterization |
| CN114529436A (en) * | 2022-02-10 | 2022-05-24 | 珠海读书郎软件科技有限公司 | Knowledge point mastery degree evaluation method, device and medium |
| CN114238613A (en) * | 2022-02-22 | 2022-03-25 | 北京一起航帆科技有限公司 | Method and device for determining mastery degree of knowledge points and electronic equipment |
| CN116308935A (en) * | 2023-05-19 | 2023-06-23 | 湖南师范大学 | Learning early warning method and device based on multi-feature modeling and multi-level evaluation |
| CN116308935B (en) * | 2023-05-19 | 2023-08-08 | 湖南师范大学 | Learning early warning method and device based on multi-feature modeling and multi-level evaluation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112527821A (en) | Student bloom mastery degree evaluation method, system and storage medium | |
| Raga et al. | Early prediction of student performance in blended learning courses using deep neural networks | |
| CN111159419B (en) | Knowledge tracking data processing method, system and storage medium based on graph convolution | |
| Chaudhury et al. | An empirical study on attribute selection of student performance prediction model | |
| Intisar et al. | Classification of online judge programmers based on rule extraction from self organizing feature map | |
| CN110366735A (en) | Analyze method, equipment and the computer program of data | |
| CN113283488B (en) | Learning behavior-based cognitive diagnosis method and system | |
| Stanković et al. | Artificial neural network model for prediction of students’ success in learning programming | |
| Maslovskyi et al. | Adaptive test system of student knowledge based on neural networks | |
| Moon et al. | Rich representations for analyzing learning trajectories: Systematic review on sequential data analytics in game-based learning research | |
| CN114117033B (en) | Knowledge tracking method and system | |
| CN115205072A (en) | Cognitive diagnosis method for long-period evaluation | |
| Koong et al. | The learning effectiveness analysis of JAVA programming with automatic grading system | |
| Song et al. | A Hybrid CNN-LSTM Model for Video-Based Teaching Style Evaluation | |
| Al Karim et al. | Evaluating the performance of ID3 method to analyze and predict students' performance in online platforms | |
| Sun et al. | Predicting and Analyzing College Students’ Performance Based on Multifaceted Data Using Machine Learning | |
| Mustaffa et al. | Personality predictive analysis based on artificial neural network | |
| Villa-Torrano et al. | Early prediction of students' efficiency during online assessments using a Long-Short Term Memory architecture. | |
| Hirose | Analysis of Fluctuations of Ability Estimates in Testing | |
| Liao et al. | Predicting learners' multi-question performance based on neural networks | |
| Murray | Using decision trees to understand student data | |
| Deviatiarova et al. | Analyzing Proficiency Patterns in Test Results Using Clustering and Augmentation Algorithms | |
| CN117540104A (en) | Learning group difference evaluation method and system based on graph neural network | |
| Durai | PREDICTIVE ANALYTICS USING DEEP NEURAL NETWORKS | |
| Salem et al. | Data mining techniques for classifying and predicting Teachers’ performance based on their evaluation reports |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 116000 room 206, no.8-9, software garden road, Ganjingzi District, Dalian City, Liaoning Province Applicant after: Neusoft Education Technology Group Co.,Ltd. Address before: 116000 room 206, no.8-9, software garden road, Ganjingzi District, Dalian City, Liaoning Province Applicant before: Dalian Neusoft Education Technology Group Co.,Ltd. |
|
| CB02 | Change of applicant information |